The potentially reversible myocardial depression of sepsis is well documented in humans and animals by radionuclide scans and intravascular catheter techniques. The mechanism of sepsis-induced myocardial depression remains incompletely understood. Sepsis induced myocardial dysfunction cannot be explained by inadequate myocardial oxygen supply or insufficient myocardial high-energy synthetic capabilities. Investigators have postulated a myocardial depressant factor of sepsis, but the mechanisms by which bacteria, their toxins, and host cytokines disturb normal cardiac function remains unknown. Proinflammatory mediators have been implicated in the pathogenesis of congestive heart failure and the myocardial depression of sepsis. There is also electron microscopic evidence of diffuse abnormalities of the cardiac microvasculature characterized by endothelial cell swelling and nonocclusive intravascular fibrin deposition in septic animals. One can postulate that bacterial toxins and the induced host proinflammatory response disrupt the integrity of the myocardial microvasculature and subsequently injure the myocytes resulting in myocardial functional depression. Similar to congestive heart failure, the heart adapts and maintains stroke volume through a remodeling mechanism resulting in a reversible ventricular dilatation. The concept of ventricular dilatation in sepsis remains controversial. Sepsis studies using echocardiography to assess ventricular volumes have confirmed in humans and animals the depression of LV ejection fraction but not the LV dilatation. The purpose of this study is to better define systolic and diastolic abnormalities of the heart during sepsis and to determine if the sepsis-induced proinflammatory response results in a cardiac microvascular injury that can lead to myocardial functional depression. We will quantify the changes in cardiac function using both invasive hemodynamic measurements and noninvasive cardiac magnetic resonance imaging (MRI). The data from the invasive measurements will be correlated with the non-invasive MRI data in order to develop an approach suitable for future human studies. Furthermore, this study is designed to definitively determine if sepsis-induced myocardial depression is associated with microvascular flow abnormalities and LV dilatation.